17372021070

Committed 01 Sep 2025 08:26AM UTC coverage: 93.704% (-0.2%) from 93.906%

Build # 17372021070

Build Type

Pull #431

github

Committed by

web-flow

Commit Message

Merge 7d2c68dac into 68f21c1d6

Pull Request Pull Request #431: Product kernel explainer

Run Details

180 of 203 new or added lines in 14 files covered. (88.67%)

4 existing lines in 2 files now uncovered.

5090 of 5432 relevant lines covered (93.7%)

0.94 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

74.07

/src/shapiq/explainer/product_kernel/conversion.py

"""Functions for converting scikit-learn models to a format used by shapiq."""

from __future__ import annotations

from typing import TYPE_CHECKING

from shapiq.explainer.product_kernel.base import ProductKernelModel

if TYPE_CHECKING:
    from shapiq.typing import Model


def convert_svm(model: Model) -> ProductKernelModel:
    """Converts a scikit-learn SVM model to the product kernel format used by shapiq.

    Args:
        model: The scikit-learn SVM model to convert. Can be either a binary support vector classifier (SVC) or a support vector regressor (SVR).

    Returns:
        ProductKernelModel: The converted model in the product kernel format.

    """
    X_train = model.support_vectors_
    n, d = X_train.shape

    if hasattr(model, "kernel"):
        kernel_type = model.kernel
        if kernel_type != "rbf":
            msg = "Currently only RBF kernel is supported for SVM models."
            raise ValueError(msg)
    else:
        msg = "Kernel type not found in the model. Ensure the model is a valid SVC or SVR."
        raise ValueError(msg)

    return ProductKernelModel(
        alpha=model.dual_coef_.flatten(),
        X_train=X_train,
        n=n,
        d=d,
        gamma=model._gamma,  # noqa: SLF001
        kernel_type=kernel_type,
        intercept=model.intercept_[0],
    )  # TODO (IsaH57): check if gamma is always needed or just when rbf is used (Issue #425)


def convert_gp_reg(model: Model) -> ProductKernelModel:
    """Converts a scikit-learn Gaussian Process Regression model to the product kernel format used by shapiq.

    Args:
        model: The scikit-learn Gaussian Process Regression model to convert.

    Returns:
        ProductKernelModel: The converted model in the product kernel format.

    """
    X_train = model.X_train_
    n, d = X_train.shape

    if hasattr(model, "kernel"):
        kernel_type = model.kernel_.__class__.__name__.lower()  # Get the kernel type as a string
        if kernel_type != "rbf":
            msg = "Currently only RBF kernel is supported for Gaussian Process Regression models."
            raise ValueError(msg)
    else:
        msg = "Kernel type not found in the model. Ensure the model is a valid Gaussian Process Regressor."
        raise ValueError(msg)

    return ProductKernelModel(
        alpha=model.alpha_.flatten(),
        X_train=X_train,
        n=n,
        d=d,
        gamma=(2 * (model.kernel_.length_scale**2)) ** -1,
        kernel_type=kernel_type,
    )

1	"""Functions for converting scikit-learn models to a format used by shapiq."""
2
3	from __future__ import annotations	1✔
4
5	from typing import TYPE_CHECKING	1✔
6
7	from shapiq.explainer.product_kernel.base import ProductKernelModel	1✔
8
9	if TYPE_CHECKING:	1✔
NEW 10	from shapiq.typing import Model	×
11
12
13	def convert_svm(model: Model) -> ProductKernelModel:	1✔
14	"""Converts a scikit-learn SVM model to the product kernel format used by shapiq.
15
16	Args:
17	model: The scikit-learn SVM model to convert. Can be either a binary support vector classifier (SVC) or a support vector regressor (SVR).
18
19	Returns:
20	ProductKernelModel: The converted model in the product kernel format.
21
22	"""
23	X_train = model.support_vectors_	1✔
24	n, d = X_train.shape	1✔
25
26	if hasattr(model, "kernel"):	1✔
27	kernel_type = model.kernel	1✔
28	if kernel_type != "rbf":	1✔
29	msg = "Currently only RBF kernel is supported for SVM models."	1✔
30	raise ValueError(msg)	1✔
31	else:
NEW 32	msg = "Kernel type not found in the model. Ensure the model is a valid SVC or SVR."	×
NEW 33	raise ValueError(msg)	×
34
35	return ProductKernelModel(	1✔
36	alpha=model.dual_coef_.flatten(),
37	X_train=X_train,
38	n=n,
39	d=d,
40	gamma=model._gamma, # noqa: SLF001
41	kernel_type=kernel_type,
42	intercept=model.intercept_[0],
43	) # TODO (IsaH57): check if gamma is always needed or just when rbf is used (Issue #425)
44
45
46	def convert_gp_reg(model: Model) -> ProductKernelModel:	1✔
47	"""Converts a scikit-learn Gaussian Process Regression model to the product kernel format used by shapiq.
48
49	Args:
50	model: The scikit-learn Gaussian Process Regression model to convert.
51
52	Returns:
53	ProductKernelModel: The converted model in the product kernel format.
54
55	"""
56	X_train = model.X_train_	1✔
57	n, d = X_train.shape	1✔
58
59	if hasattr(model, "kernel"):	1✔
60	kernel_type = model.kernel_.__class__.__name__.lower() # Get the kernel type as a string	1✔
61	if kernel_type != "rbf":	1✔
NEW 62	msg = "Currently only RBF kernel is supported for Gaussian Process Regression models."	×
NEW 63	raise ValueError(msg)	×
64	else:
NEW 65	msg = "Kernel type not found in the model. Ensure the model is a valid Gaussian Process Regressor."	×
NEW 66	raise ValueError(msg)	×
67
68	return ProductKernelModel(	1✔
69	alpha=model.alpha_.flatten(),
70	X_train=X_train,
71	n=n,
72	d=d,
73	gamma=(2 * (model.kernel_.length_scale2)) -1,
74	kernel_type=kernel_type,
75	)

mmschlk / shapiq / 17372021070

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous